Gulf War Illness: Challenges Persist

Abstract

It has been more than 20 years since the United States and coalition forces entered Kuwait and Iraq. Actual combat was of remarkably short duration: less than 1 week of sustained ground activity and 6 weeks of air missions. Thus, it was surprising when approximately 200,000 returning US veterans were affected by a chronic multi-symptom illness that came to be known as Gulf War Illness (GWI). There were many challenges in investigating GWI, not least of which was that it took several years before the condition was officially taken seriously. There were multiple exposures to potentially causal agents on and off the battlefield, but these exposures were documented incompletely if at all, leaving epidemiologists to rely on self-report for information. In the past 2 years, significant controversy has arisen over the future directions of the field. Despite these challenges, several studies have implicated exposure to acetylcholinesterase inhibitors such as pyridostigmine bromide in the genesis of the condition. The story of GWI can inform research into other conditions and guide future work on veterans' health.

INTRODUCTION

In August 2, 1990, Iraqi forces invaded Kuwait on the pretense that the two countries had been unified in the days of the Ottoman Empire and thus should be reunited. The real reasons for the invasion were more complex, involving economic and sociopolitical factors. Kuwait's military capabilities were small in comparison to those of Iraq and they were rapidly overwhelmed. Within 1 day, the majority of Kuwait was occupied by Iraqi troops. The President of Iraq, Saddam Hussein, had alienated many neighboring countries and there was concern that the conflict could extend into Saudi Arabia and beyond. Ultimately, a coalition force from 34 countries was assembled under the leadership of the United States to liberate Kuwait and drive Iraqi forces back within their borders. Operation Desert Storm began on January 16, 1991, with a 6-week bombing campaign followed later by invasion of ground forces. The ground war in Kuwait lasted less than 1 week. Operation Desert Storm successfully drove Iraqi troops out of Kuwait and Iraq signed the United Nations resolution officially ending the war on April 6, 1991. Almost 700,000 US troops were deployed during the 1991 Gulf War. Both escalation and de-escalation were rapid, with only 50,000 US troops still deployed in June of 1991 (1).

Despite the brevity of the war, almost one fourth of troops experienced a chronic, multi-symptom illness after deployment (2). Commonly known as Gulf War Illness (GWI), the condition was concerning because of the large numbers of cases and the inability of medical science to pinpoint a cause. More than 20 years later, symptoms persist in many veterans of the Gulf War. The purpose of this article is to review the research and controversies surrounding GWI and to emphasize the difficulties encountered by researchers and patients.

CLINICAL PRESENTATION AND DEFINITIONS

GWI is an unexplained, multi-symptom illness occurring in veterans of the 1991 Gulf War (2). Symptoms vary, but commonly include fatigue and difficulty with memory and/or concentration. Gastrointestinal symptoms, respiratory complaints, pain, and rashes also are noted in some patients. Symptoms appeared during or shortly after deployment, and often did not improve over time (3). Routine clinical laboratory tests are unremarkable and the search for a biomarker has been unsuccessful.

There is no uniform case definition of GWI. Most studies have used the Kansas definition or the definition put forth by investigators for the Centers for Disease Control and Prevention (CDC). The Kansas definition includes symptoms in three of the following six areas: fatigue/sleep, somatic pain, neurologic/cognitive/mood, gastrointestinal symptoms, respiratory symptoms, and skin symptoms (4). The Kansas definition excludes cases that have established diagnoses to explain symptoms. The CDC paper defined GWI more generally as having at least one chronic symptom from two of the following three areas: mood/cognition, fatigue, and musculoskeletal (5). In 2014, the Institute of Medicine (IOM) (6) reported that their expert panel found merits to both the Kansas and CDC definitions and recommended that the US Department of Veterans Affairs (VA) use either one depending on the need for a more specific or more general definition, respectively. A third definition has been developed and was recently validated (7), describing three subgroups or “variants” of GWI.

The VA does not endorse a specific definition of GWI, and does not refer officially to the condition by that name, often using the term “chronic multi-symptom illness.” The IOM has recommended that the VA adopt the term GWI (6), but this has not been done. However, the VA will consider Gulf War veterans for disability coverage if they have a chronic, “medically unexplained illness” that is independently verified (8). This definition is broad and includes GWI, chronic fatigue syndrome, fibromyalgia, and functional gastrointestinal disorders. Moreover, it does not appear that large numbers of Gulf War veterans have received disability coverage based on these definitions.

EPIDEMIOLOGY

The prevalence and types of symptoms vary among individuals and some non-deployed veterans also have symptoms consistent with GWI. Most studies found the excess rate of GWI in deployed versus non-deployed veterans was 25% to 30%, indicating that 175,000 to 210,000 soldiers were affected (2). Both men and women were affected, although some studies found that women were disproportionately affected (9). However, due to the male predominance in veteran populations, the majority of cases occurred in men.

POTENTIAL EXPOSURES AND CAUSES

A variety of exposures have been investigated as potential causes of GWI (Table 1). For example, Iraqi forces set fire to Kuwaiti oil wells as they retreated causing days to weeks of heavy air pollution, at times turning the air black. Depleted uranium, an extremely dense material, was used in tank armor and some weapons. Due to fears of biological weapons, anthrax vaccine was used liberally among troops. Some troops were exposed to infectious diseases that would be considered rare in the United States including schistosomiasis and leishmaniasis. Concerns about exposure to sand flies that spread leishmania and other insects resulted in some troops using insecticides, treating uniforms with pesticides and even wearing flea collars. Thus, organophosphate exposure was high in some groups.

TABLE 1.

Potential Exposures

Smoke from oil well fires

Depleted uranium

Infectious diseases

Anthrax vaccine

Organophosphate pesticides/insecticides

Sarin nerve gas

Mustard gas

Pyridostigmine bromide

Chemical agent resistant coating

Combat stress

Other exposures

It was well known that Iraq possessed chemical weapons and exposure to mustard gas or sarin nerve gas was a significant concern. Low-level exposures may have occurred (10), but were difficult to document. However, the only large-scale release occurred when allied troops destroyed a munitions depot at Khamisiyah containing nerve gas. Because of concerns about nerve gas, troops were issued pills containing pyridostigmine bromide (PB). PB was to be taken before anticipated exposure to sarin nerve gas. Both sarin gas and PB bind and inactivate acetylcholinesterase, but sarin binds irreversibly whereas PB binding is reversible. In theory, pretreatment with PB would protect PB-bound acetylcholinesterase from subsequent binding to sarin, although this theory has not been tested in humans.

RESEARCH INTO CAUSES

Investigation of GWI has been significantly hampered by the delay in its recognition and the lack of consensus on a case definition. Exposures in theater were not recorded, so there is no systematic database to analyze. For this reason, epidemiologic studies have relied heavily on self-report and thus are open to the criticisms inherent with this methodology. Nonetheless, interesting associations have been identified suggesting potential causes of GWI.

Most studies enrolled a cohort of veterans, comparing those who had GWI to those who did not or comparing deployed and non-deployed personnel. Univariate analysis often revealed significant associations between exposures and GWI, but multivariate analysis adjusting for multiple effects identified far fewer associations. For example, the CDC surveyed a unit of the Air National Guard that had a high rate of symptomatic personnel and three other military air units (11). Of the 1,163 participants enrolled, analysis was done on 1,002 participants who had complete data, of whom 45.8% were defined as having a chronic multi-symptom illness consistent with GWI. Univariate analysis identified multiple associations with illness (11). Multivariate analysis adjusting for age, sex, smoking, and rank identified three factors as being independently associated with GWI: use of PB (odds ratio [OR]: 2.9, 95% confidence interval [CI]: 1.4–6.1), regular use of insect repellant (OR: 2.4, 95% CI: 1.3–4.5), and belief that biological weapons (OR: 3.5, 95% CI: 1.7–6.9) or chemical weapons had been used (OR: 2.3, 95% CI: 1.5–3.3).

Studies used varying definition of GWI, sometimes including diagnoses of chronic fatigue syndrome, post-traumatic stress disorder (PTSD), inflammatory bowel disease, and multiple chemical sensitivity syndrome. For example, a large cohort study in 1997 surveyed US Navy Seabees who were active during the Gulf War era, assembling a cohort of 3,831 Gulf War Seabees (9). The case definition included people who reported any of the above conditions or who had at least 12 of 33 medical symptoms. As expected, Seabees deployed to the Gulf had a higher incidence of this broadly defined chronic multi-symptom illness compared to non-deployed or elsewhere-deployed Seabees. Among deployed Gulf War Seabees, multivariate analysis found a weak (odds ratio [OR] < 2.0) but significant (P < .05) association between 12 potential exposures and illness.

Deployment experiences varied among veterans with some being stationed in support areas and others stationed in the Iraq-Kuwait theater. A 2012 study by Steele et al (12) found that deployment experiences were important in GWI. Specifically, multivariate analysis showed that GWI in veterans who had been deployed in Iraq or Kuwait was associated with the use of PB (OR: 3.5, 95% CI: 1.7–7.4), use of pesticides on skin (OR: 2.07, CI: 1.06–4.05) and being in the vicinity of an exploding Scud missile (OR: 3.1, 95% CI: 1.5–6.1). Veterans deployed in support areas did not have these associations. Direct exposure to combat was not associated with GWI in this study.

There is no animal model for GWI. However, mice treated with PB and insecticides such as permethrin show problems with memory and cognition, especially if stressed (13–15). Loss of neurons, reduced neurogenesis, activation of microglia, and other findings support neurocognitive testing (13–15).

Large studies have not revealed distinct abnormalities on physical examination. However, a single smaller study found autonomic deficits in selected areas including reduced circadian variation in high frequency heart rate variability and measures of sudomotor function in the foot (16). Similarly, there is no biomarker for GWI and standard laboratory testing has not revealed abnormalities. However, the results of small studies show abnormalities in selected areas including mitochondrial dysfunction and neuroimaging (17–19). Further studies are needed to replicate and enlarge on these findings.

In 2008, the relevant research was summarized by the congressionally mandated Research Advisory Committee on Gulf War Veterans' Illnesses (2). The group concluded that GWI was a serious condition that differed from PTSD and other stress-related syndromes. The Committee identified two exposures during deployment that were “strongly and consistently” associated with GWI including PB use and pesticide use during deployment, creating a “persuasive case” for causality. They cited the weight of accumulated epidemiologic studies, animal studies of the agents, and dose-response effects in coming to this conclusion. The Committee went further and stated that there was little likelihood that depleted uranium, anthrax vaccine, hydrocarbons, air pollutants, or infectious diseases were causally associated with GWI. They concluded that the evidence could not rule out associations with oil well fire exposures and multiple vaccinations. Finally, they concluded that research funding was inadequate to the task of advancing understanding of GWI.

In 2009, the IOM published its own statement (1). They agreed that there was sufficient evidence that deployment to the Gulf War was associated with a chronic multi-symptom illness. They also found an association of deployment with chronic fatigue syndrome, anxiety disorder, and functional gastrointestinal illness. Finally, they concluded that deployment played a causal role in the occurrence of PTSD. The IOM was not charged with reviewing the literature on potential exposures and GWI, but submitted an appendix commenting on the conclusions of the Research Advisory Committee. The IOM stated that they did not feel that the evidence implicating PB and pesticide use was conclusive enough to assign causality. In particular, they cited the lack of objective exposure data and the need to rely on self-report as barriers to establishing causality.

War creates significant stress on soldiers and there were undoubtedly veterans who experienced PTSD (1). However, PTSD did not explain the vast majority of cases of GWI (2). In part, this was due to the relative brevity of combat. Yet, stress has multiple physiological effects and has been suggested as a sensitizing condition that might predispose soldiers to conditions such as GWI in the presence of appropriate exposures (20).

Chronic multi-symptom illness that begins during/after deployment has been noted in the recent Iraq and Afghanistan conflicts, but at much lower levels than were observed in the Gulf War (21,22). Indeed, most if not all wars have been associated with medically unexplained symptoms in returning veterans. Using war pension files, Jones et al reviewed the medical and military records of a random sample of 1,856 British veterans who served in wars between 1854 and 1991 and had war syndromes compatible with unexplained multi-symptom illnesses (23). The results showed that soldiers from all wars were affected, although the terms used to describe the conditions changed over time and included shell shock, disordered action of the heart, neurasthenia, and psychoneurosis. Symptoms most commonly included difficulty completing tasks, fatigue, dyspnea, anxiety, and weakness. The comparison wars in the Jones article lasted years at a time, whereas the exposure to the theater of operations in the Gulf War was measured in weeks or months.

TREATMENT

Treatment studies have been problematic, partially because of the difficulty in defining the illness and identifying clear subpopulations that might benefit. In 2013, the IOM concluded that no treatment had been found to be effective and recommended supportive care that is individualized. Based on studies of other multi-symptom illnesses, they suggested consideration of cognitive behavioral therapy and selective serotonin re-uptake inhibitors (12).

DISCUSSION

Gulf War Illness is a chronic multi-symptom condition associated with deployment during the Gulf War. Although other wars have been associated with multi-symptom illnesses, GWI is unique in that it affected such a large number of returning soldiers and that it occurred after a relatively brief exposure to combat.

The study of GWI has been complicated by many factors. The initial response to ill veterans was to ignore the problem, imply that they were somehow not as tough as veterans of previous wars or generations, and to resist calls for funding to study the problem. This resulted in significant delays in research. Objective data on exposures were not available, forcing studies to rely on self-report with its attendant biases. The US Department of Defense has pledged to improve tracking of potential exposures and recommendations have been made on this topic (24). The lack of a standard case definition for GWI has been particularly problematic because some otherwise well-done studies have used vague definitions causing confusion about results. In addition, symptoms of GWI overlap with those of other syndromes, making it likely that other conditions are inadvertently included in studies of veterans.

For these reasons, it is perhaps not surprising that the Research Advisory Committee (RAC) and the IOM came to different conclusions when reviewing the data.

The RAC emphasized the preponderance of the evidence in nominating PB and insecticides as potential causal factors. The IOM emphasized the difficulties faced by researchers which affected the strength of study design and injected more doubt into the conclusions of the epidemiological studies. Although both the RAC and IOM concur that GWI is a chronic multi-symptom condition associated with deployment, the VA has never fully accepted the idea and avoids using the term Gulf War Illness. In its original advocacy role, the RAC was critical of the VA's general approach and funding strategy for GWI research. Recently, the charge, membership, and reporting relationship of the RAC was changed, leading some to suggest these moves were made in response to the criticism (25, 26).

More than two decades after the Gulf War, GWI remains a controversial topic with no established treatment to relieve those who are affected. Veterans continue to be affected by GWI and merely deploring the problems inherent in studying the condition will not serve them or advance the field. Acetylcholinesterase inhibitors such as PB and insecticides cannot be overlooked as potentially causal agents of GWI and should continue to be studied. Investigation may help current and future veterans.

DISCUSSION

Moore, New York: When you went to medical school you probably didn't think you would be presenting a talk on Gulf War Illness. How did you get from medical school to studying Gulf War Illness?

Nettleman, Sioux Falls: I am an epidemiologist and an infectious disease physician. Those were my qualifications. That was why Congress tapped me to the Gulf War committee. At the time, people thought it might be an infectious disease. I was there to actually provide expertise on infectious disease and epidemiology to the Congressional Committee. That is how I got into this in the beginning. This is not the major focus of my research. None of these papers, with the exception of the reports, have my name on them. I do find it a very fascinating field, and it appears, unfortunately, to be one that is evergreen.

Olds, Riverside: There are actually two other interesting stories related to that war. One is, there is a new syndrome with leishmaniasis that was discovered as a result of that — that isn't the cause, by the way, of Gulf War Syndrome — a clinical syndrome that was previously unknown until we exposed large numbers. The other intriguing story is that all of the veterans during the Gulf War were given mefloquine since this was a “malaria endemic area.” For those of us that are not tropical disease specialists, mefloquine is unnecessary there since there is no chloroquine-resistant malaria; you could have used chloroquine. But mefloquine crosses the blood brain barrier. In fact you're specifically instructed not to do dangerous things like drive cars, etc. I guess using M-14s is not in that category. But we are using the drug on our troops that is completely unnecessary because they had only one-size fits all. They should have been using chloroquine, and they continue to use mefloquine in the Gulf theater. It is interesting that these wars have brought out a lot of unexpected issues in healthcare that we probably should be more responsive to.

Nettleman, Sioux Falls: I would argue too that we need a better recording system and more transparency with the recording system, so that we could actually do some of those studies. As you point out, there really isn't any malaria in the first Gulf War regions. The second Gulf War went further into Iraq. There were some minor areas, and it's still not a very big issue. When we give vast amounts of medications that appear to us to be safe, using them in small amounts in relatively healthy people in calm situations; it's different when you give it to people under high stress conditions, I think.

Sanyal, Richmond: If you change the balance between sympathetic and parasympathetic activity, you have autonomic imbalance. Is there autonomic dysfunction in these patients? Is there any evidence of neuroinflammation that might explain the chronic fatigue−like symptoms in this population?

Nettleman, Sioux Falls: There are small studies that have been done that have shown some evidence of autonomic dysfunction, pseudomotor dysfunction, and other things; also, mitochondrial dysfunction. They are very small studies. There is a series of these small studies — all of which that show things that are very interesting — but are so small that I couldn't tell you. There is no simple test that you could do of the autonomic nervous system like you do in a doctor's office that would identify these Gulf War veterans. And I should mention right now, so far there doesn't appear to be a difference in mortality rates in Gulf War Illness veterans, although they have a higher use of outpatient services. The same is true of inflammation. There are some small studies showing inflammatory markers; they aren't very sophisticated studies. There is brain repository that has been started to try to get at this, and it may be that we'll know the answer in the future.

Hochberg, Baltimore: We still see a lot of patients with Gulf War Syndrome at the arthritis clinic at the Baltimore VA Medical Center, and it's very challenging for treatment. Dr Crofford commented in terms of the challenges in treating people with chronic pain syndromes. It's very difficult to manage patients. So I wonder if you have any insight into the treatment paradigm for patients with Gulf War Syndrome.

Nettleman, Sioux Falls: The Institute of Medicine put out a couple of years ago a review of the topic and came up with the fact that you should do supportive therapy, by which they mean general things like cognitive behavior therapy and other interventions because nothing else has been shown to work. There have been many trials. There was a Co-Q enzyme study that I think was more in the interest of the people doing the study. They were so interested in it and so wanted it to work and they were sure it was going to, and it just kind of flopped. There hasn't been any therapy that has worked. In my opinion, there was damage done and you can't undo that damage. There probably is some lingering malfunction that could be helped with the cognitive therapy and other things. But I don't think we are going to find a magic bullet that ever is going to fix this syndrome… and that is a lot people, as you say. Most of these people are employed, they try to work through it, they are young people, they are in their 30s now.